Computer systems are currently in wide use. Some computer systems are relatively large and may include, for instance, thousands of different forms (e.g. electronic fillable documents) and other types of elements. Such computer systems are often customized (some heavily so) before they are deployed in a given implementation. Some large computer systems may include many thousands of different forms, each of which may have different controls and other user interface elements. Some example systems also include execution logic as well as workflows that allow end users to access the system and perform a set of activities, or tasks, in order to carry out their duties in conducting a particular operation.
When such a computer system is deployed in a specific environment, it is common for the system to be customized to meet the functional requirements of the particular end user. By way of example, different customers may wish to have different form controls on a given form that represents a particular entity. In addition, different customers may also wish to have different fields, execution logic, or other items on a particular report form. Thus, it can be seen that a given computer system may be highly customized so that it meets the requirements of a particular end user for which the system is implemented.
When the base system is authored (before it is customized) it may also include a number of different tests that can be run to determine whether the system is working properly. Additionally, as a developer customizes or continues to develop on the base system, the developer may generate additional tests as well. Further, once the development is complete, it is also important to ensure that a system is functioning properly so that it will effectively handle real data.
In some cases, synthetic data is used to test the system. Generating a large volume of synthetic data for such a system is a complex and long-running process that usually involves many different entities and complex logic. It can be difficult to generate the synthetic data rapidly without introducing any errors. It is also difficult to maintain the relational integrity if the data generation is done in parallel.
The discussion above is merely provided for general background information and is not intended to be used as an aid in determining the scope of the claimed subject matter.